ClawHub
Back to skill

Security audit

Ethical Acquisition

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed planning and approval-gating workflow for small acquisition actions, with strong limits before any external contact or account change.

Install only if you want an agent to help draft acquisition outreach or listing assets. Before any send, publish, spend, or account change, verify the exact payload, destination, account, cost, rollback path, consent basis, and platform rules yourself.

A.I.G

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
75% confidence
Finding

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Content

Scanner excerpt · SKILL.md (reported line 22)May include surrounding context.

md
`Payable Test Plan`, including the buyer, payment condition, scope, evidence
label, cost and work ceilings, delivery authority, stop condition, and approval
checkpoint. Run Authorization permits private channel analysis, drafting, and
artifact creation without asking for another approval. Return to
`moneyprinter`, `offer-engine`, or `payable-test` when the route materially
changed or the offer cannot be presented honestly. Run Authorization never
permits a buyer-facing or account-mutating action.

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
75% confidence
Finding

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Content

Scanner excerpt · SKILL.md (reported line 118)May include surrounding context.

md
decline, and no unverified performance claim. In `Approval checkpoint`, include
the exact payload, destination, expected cost, rollback or correction path,
`Approver authority status`, `Third-party data status`, and
`Regulated-review status`. Include an explicit yes/no approval request only
when every required prerequisite passes; otherwise print
`Execution blocked — prerequisite unresolved`. Keep the asset staged.
In `Private fulfillment preparation`, include the deliverable workflow,

Static analysis

No suspicious patterns detected.